August 23, 20196 yr I'm having trouble isolating what is causing IOWAIT to consume huge chunks of CPU every 30~45 seconds. I first thought it was the new PLEX docker, then I thought it was "Folder Caching" app. Disabling either of these allows the CPU to remain normal for a while, but later it's utilization goes up again. Also, yesterday with the command prompt I deleted a directory (share) and that lowered the CPU utilization for a while ?!? The screen grab shows the CPU during a file transfer from the cache drive to the array. This file transfer goes a bit slower than normal. I tried copying it twice from cache to the array and both times it slowed down and both times it wrote to a different data disk. This makes me think I may have an issue with either the cache disk or parity disc? I am puzzled I've changed a lot on the server lately including swapping out an 8TB for a 10TB cache drive, installing NORCO 3-5 drive pods, and a few other things... Attached are a screen grab and diagnostics. Any help would be appreciated. Thanks, craigr unraid-diagnostics-20190823-1906.zip Edited August 23, 20196 yr by craigr
August 23, 20196 yr There’s already an open bug report. Revert to v6.6.7. https://forums.unraid.net/bug-reports/stable-releases/67x-very-slow-array-concurrent-performance-r605/
Archived
This topic is now archived and is closed to further replies.